Home | Databases | WorldLII | Search | Feedback New Zealand Yearbook of International Law

Mushkat, M; Mushkat, R --- "International Law and Game Theory: A Marriage of Convenience or Strange Bedfellows?" [2005] NZYbkIntLaw 4; (2005) 2 New Zealand Yearbook of International Law 101

INTERNATIONAL LAW AND GAME THEORY: A MARRIAGE OF CONVENIENCE OR STRANGE BEDFELLOWS?

Miron Mushkat, Roda Mushkat^[*]

I. Introduction

International law cannot be portrayed as an inertia-driven academic discipline. It has evolved broadly in tandem with changes in the international system, albeit at a somewhat slower pace, and not invariably in a smooth fashion. It has displayed openness to ideas emanating from other academic sources, although again this statement is subject to a number of qualifications. The relationship between international law and international relations, an academic discipline exposed to a wide away of theoretical and methodological currents, has been particularly close, and productively so.

That said, international law has arguably exhibited greater moderation regarding conceptual and technical imports from the social and natural sciences than the domestic branches of the law. There is ample evidence to substantiate this argument. Law and economics, and law and society, have for some time been established as key components of legal curricula and have loomed large on the legal research agenda. Yet, the pedagogical and analytical efforts in both areas have been heavily tilted toward the domestic end of the legal spectrum. The domestic branches of the law have also been decidedly more active in borrowing from fields such as mathematics and statistics.

This pattern has undergone some changes in recent years. Students of international law have tended to venture beyond their traditional domain more freely than in the past. By the same token, they have increasingly been approaching their subject from a broad multi-disciplinary -- at times, even inter-disciplinary -- perspective, incorporating insights and tools from other academic sources. The social sciences have loomed large on this evolving research agenda, but the natural sciences have not been overlooked. It should also be noted that the “paradigm shift” has not been confined to a single institution (e.g., Yale University) or one country (e.g., the United States).

It would be inappropriate to stretch the argument too far in this direction. To suggest that a paradigm shift has occurred may be an overstatement. Similarly, certain institutions have provided a more fertile ground in this respect than others. And the diffusion of “new” ideas has not been even across countries/regions. Perhaps most importantly, some parts of international law have proved relatively open to external influences, but certain parts have displayed greater inertia. This is a common experience in cross-disciplinary contexts and should not be viewed as an international law-related aberration.^[1]

The “formation” and “maintenance” of international law, both of the customary and treaty varieties, may thus be singled out as “areas” where the stretching of theoretical and methodological boundaries has been pursued in a relatively liberal fashion in recent years. Economic concepts and techniques have had a particularly strong appeal, providing inspiration for scholars with a penchant for utilitarian logic and rigorous manipulation of inter-related variables. Game theory has often been invoked in such circumstances in order to generate explanatory insights (i.e., as a tool of description) or offer normative observations (i.e., as a tool of prescription).

The maintenance of international law -- or, to express it in more familiar terms, international legal compliance -- has furnished a particularly convenient entry point for practitioners of game theory seeking new intellectual adventures. The subject has long been examined within a framework emphasising the “honourable intentions” of actors in the international arena and arguably, has had to be exposed to a utilitarian critique with clear “sceptical” undertones. International legal compliance has also been viewed as a largely “static” phenomenon and game theory, which explores multi-party interactions over time, has provided a much-needed “dynamic” dimension. Overall, the impression is that the encounter between international law and this branch of economics/mathematics has been a productive one.

The negative side of the picture should not be overlooked, however. The purpose of this paper is to highlight that side, but not to a point of obscuring the positive aspects. The focus is on international legal compliance because this is the field to which game theorists have most strongly been drawn, yet the conclusions apply to international law as a whole. The paper consists of three parts. First, a game-theoretic type of a shift in the study of international legal compliance is outlined. An illustration is then offered featuring Hong Kong in the China context. This is followed by a discussion of the limitations of game theory as a tool of international legal analysis. The third part constitutes the core of the paper.

II. Compliance Puzzle

Students of international law have traditionally assumed a high degree of compliance on the part of players -- predominantly states -- bound by international rules. This position has manifested itself in a variety of analytical forms. The “management approach” is a relatively sophisticated variant, displaying greater coherence than the alternatives. Scholars who identify with this school of thought ascribe the general propensity of states to comply with international law to efficiency considerations, self-interest, and widely-held values. Non-compliance, when it manifests itself in a tangible form, is seldom the result of a deliberate intent to breach established agreements, but the unfortunate effect of capacity limitations and rule ambiguity. The corollary is that non-compliance should be addressed through an enlightened strategy of capacity building, rule clarification, and enhanced transparency, rather than through coercive enforcement.^[2]

Capacity constraints feature prominently in managerial-style dissection of the causes of non-compliance. As Young has observed: ‘The effectiveness of international institutions varies directly with the capacity of the governments of members to implement their provisions.’^[3] Political capacity constraints stem from the inability of signatories to induce public and private actors in the domestic arena to behave in a manner consistent with treaty obligations. Governments may fail to secure ratification, insure adherence to agreements (whether partially or across the board), or exhibit the necessary administrative skills. Economic capacity constraints come into play when financial limitations impinge on states’ ability to fulfil international commitments. Lack of sufficient resources may directly hamper compliance efforts, and macroeconomic factors may exert influence indirectly by affecting the general climate in which public and private actors operate.^[4]

Managerial theorists further argue that non-compliance may be inadvertent. For a number of reasons, both general and determined by specific circumstances, treaty language is at times imprecise and unclear, and this results in misinterpretation by states. Thus, ‘more often than not there will be a considerable range within which parties may reasonably adopt differing positions as to the meaning of the relevant treaty language.’^[5] Inadvertent non-compliance may also have its origins in the uncertainty involved in choosing the strategies required to meet treaty targets (given that different strategies can be pursued in such contexts). The problem surfaces commonly, for example, in the area of international environmental law.^[6]

This diagnosis of the causes of non-compliance translates into prescriptions for addressing violations. As Chayes and Handler Chayes contend in their seminal work on the subject: ‘If we are correct that the principal source of non-compliance is not wilful disobedience but the lack of capability or clarity or priority then coercive enforcement is as misguided as it is costly.’^[7] The attention accorded to this mode of compliance management in the academic literature does not reflect its actual use and success, maintain the two legal scholars. Or, to state it more specifically, ‘[s]anctioning authority is rarely granted by treaty, rarely used when granted, and likely to be ineffective when used.’^[8]

Instead, proponents of managerialism stress the merits of capacity building, rule clarification, and enhanced transparency as remedies for non-compliance. Whereas some political and economic capacity constraints cannot be readily alleviated through international action, deficits in technical expertise, bureaucratic effectiveness, and financial resources may be partially or even entirely eliminated via systematic capacity building.^[9] By the same token, authoritative rule interpretation in appropriate international legal settings may significantly reduce non-compliance stemming from ambiguous treaty language.^1[1]

In this line of theorising, dispute settlement is regarded primarily as clarifying common norms through interpretation and adjudication rather than serving as a channel for enforcement. However, it should be emphasised that the mechanisms of rule interpretation need not be confined to formal adjudication in international courts; informal and non-binding meditative processes may also clarify treaty rules.^[10] Enhanced transparency, the third remedy, may provide a positive reinforcement in this respect by contributing toward ambiguity reduction and fostering a sense of confidence among players in the substantive and procedural dimensions of the agreement.^[11]

In recent years, the management approach has surrendered analytical ground to its “enforcement” counterpart. The latter is firmly rooted in the political economy tradition of game theory and related branches of the social sciences (e.g., the analysis of collective action).^[12] Enforcement theorists view states -- like any other “groups” -- as rational entities that carefully assess the costs and benefits of alternative choices when making compliance decisions in international settings. Both the sources of non-compliance and the solutions to the problem emanate from the incentive structure. States opt to defect when confronted with an incentive structure in which the benefits of shirking exceed the costs of detection. By implication, strategies designed to secure compliance should seek to increase the likelihood and costs of detection through monitoring and the threat of sanctions.^[13]

The argument that states may deliberately choose not to comply hinges on the assumption that the motivation for entering an internationally “binding” compact largely centres on the signature part of the process, rather than compliance: ‘Even if a state may believe that signing a treaty is in its best interest, the political calculations associated with the subsequent decision actually to comply with international agreements are distinct and quite different.’^[14] The point is that states may attach symbolic or practical importance to the acts of participation and signing, but place a low value on the specific content of the rules, and have consequently no reservations about violating treaty provisions. The decision not to comply may also reflect conflicting priorities in that resources channelled in one direction are not available for other uses.^[15]

Scholars who identify with the enforcement approach normally stipulate that the likelihood of intentional shirking depends on the problem structure of the particular cooperative context. Collaboration, or mixed-motive, situations carry greater incentives to defect than coordination situations, where states have strong reasons to cooperate in order to avoid common aversions. In collaboration situations, states have an incentive to renege on their commitments, because they gain more from the agreement if they reap all the benefits without investing their own fair share. Since collaboration is the predominant problem structure in inter-state relations, international treaties are seldom effective without proper enforcement mechanisms.^[16]

In the absence of such mechanisms, shirking is the inevitable outcome. Monitoring and sanctions constitute the two key components of the enforcement strategy. The former enhances transparency and facilitates identification of non-complying parties/action. The latter increase the costs of shirking and render non-compliance a less appealing option. When pursued effectively, monitoring and sanctions have the potential to deter behaviour inconsistent with agreed-upon standards and induce compliance: ‘A punishment strategy is sufficient to enforce a treaty when each side knows that if it cheats it will suffer enough from the punishment that the net benefit will not be positive.’^[17]

In a paper that has attracted considerable academic attention, Downs, Rocke, and Barsoom have added an extra layer to this analytical structure. Countering arguments by managerial theorists that sanctions are seldom available and effective, they have proposed what might be termed the “depth-of-cooperation” hypothesis. This notion reflects the fact that treaties vary in the degree to which they require states to move beyond their established practices (i.e., in their “depth”). Specifically, the greater the departure from the status quo, the stronger the incentive to shirk. By the same token, the deeper the agreement, the more severe the punishments needed to support it.^[18]

According to Downs, Rocke, and Barsoom, the limited use of sanctions at the international level is largely attributable to the modest changes in state behaviour by most treaties currently in operation. International treaties codify prevailing patterns, rather than necessitate a radical adjustment in state conduct, and thus provide few incentives for non-compliance. The corollary, if one embraces the proposition, is that the management approach is based on a skewed selection of cases and suffers from endogeneity problems. It follows that, if states were to negotiate agreements requiring far-reaching behaviour shifts, enforcement would be essential for securing compliance.^[19]

III. The Hong-Kong-China Context

International legal theories need to be assessed in light of state practice, and the post-1997 Hong Kong-China relationship provides an appropriate vehicle in this respect insofar as compliance is concerned, albeit selectively so. The historical backdrop is generally familiar and requires no elaboration. Hong Kong was absorbed into the British Empire over six decades stretching to the end of the 19^th Century: Hong Kong Island was ceded by China in 1841 and 1842, Kowloon Peninsula in 1860, and the New Territories in 1898. Sovereignty over the entire area reverted to China one hundred years following the incorporation of the New Territories. Unlike the absorption, which featured a high degree of coercion, the reversion was a relatively orderly affair, although one preceded by intense bargaining. The latter took place within an elaborate legal framework embodied in the Sino-British Joint Declaration on the Question of Hong Kong. The Declaration is a legally-binding international treaty.^[2]

The formal status of the document notwithstanding, serious doubts prevailed for some time, and to some extent still do, regarding China’s willingness and ability to fulfil its obligations vis-à-vis Hong Kong. After all, China is thought to be driven by a strategic conception of legal commitments, interpreting them liberally according to changes in circumstances; the interests of its elite and Hong Kong can hardly be described as identical; and the institutional infrastructure and values underpinning it in the two jurisdictions may well be considered incompatible. The concept of “one country, two systems,” while theoretically providing a solid foundation for productive co-existence, and effectively guaranteeing politico-economic continuity beyond 1997, is thus not without problems from a practical perspective, despite its international legal origins.^[2]

To complicate matters, the Joint Declaration cannot be easily enforced. The principal mechanism for ensuring compliance is the Basic Law, the de facto Hong Kong constitution.^[21] Given the power asymmetries characterising the relationship between China and Hong Kong, one may legitimately question the effectiveness of this mechanism, however. Flagrant violation of the treaty could possibly provoke an appropriate British response at the international level. Yet, it is by no means certain that the UK government would be inclined to jeopardize British interests in China and, even if it decided to act in a resolute fashion, China would probably not submit to third-party adjudication.^[22] While the enforceability of international treaties (and the associated problem of monitoring) often constitutes a formidable challenge,^[23] the difficulties confronted by the parties in this case are particularly serious -- in some respects insurmountable.

Given the inherent fragility of this legal edifice, coupled with the perception that China’s international actions are shaped by a desire to maximize discretion rather than adhere to rules, at times opportunistically so, it is not surprising that the predicament of post-1997 Hong Kong has attracted considerable attention on the part of enforcement theorists, both in the formal and informal sense of the term, before and after the transfer of sovereignty. This section describes one such study (extending over two books), chosen for the clarity characterising the articulation of game-theoretic principles, as well as the prominence it has achieved.^[24] The purpose is not to submit the enforcement approach to an empirical test in a concrete context,^[25] but to illustrate the interplay between game theory and international law/international legal compliance in a setting where abstract concepts and complex realities converge.

The study in question employed, prior to the transfer of sovereignty, group models of politics and utilitarian logic, within an essentially game-theoretic framework, in order to forecast developments in Hong Kong beyond 1997, emphasising China’s compliance with its international legal obligations. The key features of this analytically and methodologically ambitious work are summarised below. Again, the objective is not to reflect accurately, but merely follow the development of an argument rooted in game theory, broadly defined, and applied in a manner rendering it relevant to students of international law/international legal compliance.

It should be noted that group models of politics presuppose that political activity basically consists of competition between groups over policy related issues. The pursuit of policy goals is partly motivated by a desire to maximise advantage/accumulate power, but it is not the sole driving force. Groups, and perhaps more importantly, those who lead them, aspire to promote and enforce certain policies. The two motives may be indistinguishable in practice, yet considerable emphasis is placed on separating them at the theoretical level. Competition thus does not revolve exclusively around advantage/power. Rather, it features group activity consciously directed toward policy goals.^[26]

Groups are viewed, in this context, as amalgamations of individuals displaying the same attitudes with respect to a set of policy questions. In the parlance of game theorists, a group is any entity, ranging in size from one individual to many millions, with a common preference ordering over certain issues. Given this characteristic, groups exhibit a high degree of cohesion in the policy arena (i.e., speak effectively with one voice). The corollary is that the policy preferences of any one member accurately reflect those of the entire body. Therefore, it is convenient, as well as legitimate, to assume that each group is headed by a strong leader whose own policy goals are identical to those of all the other members (the “strong leader” hypothesis).^[27]

The leader is thought to be rational. This implies that he/she rates alternatives as more or less appealing and chooses the one considered the most attractive. It follows that the first choice is preferred to the second, the second to the third, and, by extension, the first to the third. Given this pattern, game theorists describe rational decision makers as holding connected, transitive preferences. Such actors do not merely select their most preferred alternative, they also take the necessary steps to implement it.^[28]

While politics is regarded primarily as a game of competition (and coercion), this does not rule out the possibility of cooperation. Groups with similar interests form alliances, whether tactical or strategic in nature. It is interesting to note, however, that the purpose of cooperation is normally to enhance the groups’ competitive advantage. Nor is the notion of group consensus (or, to put it more emphatically, unanimity) inconsistent with the prospect of factionalism (i.e., intra-group competition). Factions emerge when there is disagreement about goals or their implementation. Another common source of factionalism is differential access to resources by group members.^[29]

Rationality, it should be emphasised, is a matter of perception. The leader is driven by “beliefs” rather than “perfect knowledge”. Indeed, rational decision makers are capable of operating in a manner that, in retrospect, appears “incorrect”, even unambiguously so. By the same token, one should not assume that two leaders, each with the same information and preferences, would inevitably opt for the same course of action. Rationality does not presuppose uniformity of behaviour. Those who embrace it merely posit that decision makers assess the costs and benefits of the alternatives open to them and choose the one that seems to yield the greatest advantage.^[30]

Utilitarian insights, grounded in mainstream microeconomic theory, are incorporated into the picture by assuming that the leader seeks to secure the largest possible net gain (or “expected utility”) based on the evaluation of the costs and benefits associated with policy choices (taking into account factors such as time schedule, constraints, appropriateness of specific decision criteria, and uncertainty). The leader is also aware of the preferences and capabilities of groups other than his/her own and proceeds accordingly. Politics, both in its competitive and cooperative forms, thus cannot be effectively dissected from the perspective of a single decision maker. Rather, one needs to resort to strategic analysis, or adopt a game-theoretic framework.^[31]

It is posited that the group and utilitarian approaches can be combined in practice to generate forecasts of a wide range of political -- including international legal -- outcomes. The information required for this purpose includes: (1) the identity of the relevant internal and external actors who may wish to influence the policy in question; (2) a set of policy alternatives that encompasses all conceivable patterns; (3) the policy preferences of each group on the issue; (4) estimates of the economic, physical (not just military), political, and social capabilities that each group may employ to influence the policy decision; and (5) estimates of the importance (salience) each group attaches to each issue, signifying the group’s interest in influencing policy outcomes.^[32]

The pre-1997 study addressing, from a game-theoretic perspective, the likelihood of China honouring its international legal commitments vis-à-vis Hong Kong tended to accord greatest attention to groups operating in the Chinese domestic political arena. The “international community” (a rather ambiguous term) and, even more so, the local elites (broadly defined, to encompass, for example, the Hong Kong foreign business community and middle-class activists) were placed on the analytical periphery. Such asymmetries, right or wrongly in specific circumstances, are a common feature of group-cantered investigations in that, inevitably, groups vary in their capabilities and the importance they attach to any particular policy issue at any particular juncture.^[33]

Within China, groups were identified on the basis of their stance regarding two key policy issues: political centralisation/decentralisation and economic reform (pro/against). “Centralisers” in Beijing were seen to be confronting “decentralisers” at the provincial level. The former were also depicted as “bureaucratic,” from the perspective of their cognitive functioning and institutional modus operandi, whereas the latter were portrayed as “entrepreneurial” in their orientation. With respect to the direction of economic strategy, a distinction was drawn between “reformers” and “conservatives.” The military was singled out as another group with a strong interest in the issues looming large on the domestic policy agenda and capable of influencing policy outcomes.^[34]

Hong Kong’s autonomy was expected to fall victim to the ruinous designs of paranoid centralisers, conservatives, and military officers. To make matters worse, decentralisers/regional leaders were likely to reinforce, rather than counter, the trend because of their desire to enhance the competitive advantages of the provinces seeking to challenge Hong Kong. One would be inclined to view such an alliance of otherwise opposed groups as an odd coalition. The pre-1997 study acknowledged the potentially problematic nature of this construct but insisted that the logic was sound with respect to the Hong Kong issue, and that the envisioned structure of interests, while very narrowly based (effectively confined to a single policy issue), would endure.^[35]

Elsewhere, it was assumed that the groups would engage in high-intensity conflict (centralisers versus decentralisers; conservatives versus reformers, etc.) rather than pursue cooperation, whether ad hoc or of the strategic variety. The pendulum would swing from one end of the policy spectrum to the other, fostering a climate of instability. In such an environment, it would prove exceedingly difficult to consistently uphold principles enshrined in the international legal and constitutional documents relating to the future of Hong Kong. Further, over time the power of those who played a meaningful role in shaping these documents on the Chinese side would materially diminish, allowing less favourably disposed groups greater leeway in the quest to curb Hong Kong’s autonomy.^[36]

The upshot would be a transformation of a society known for its laissez faire instincts and irrepressible character into one tightly controlled from “above” and drifting passively. Significant loss of freedom would be experienced at the grass roots level (people’s daily lives), in the media, at the judiciary, in academia, and even economic behaviour would be subject to socialist-style restrictions (e.g., the right to travel might not be granted unconditionally) and manipulation (e.g., contracts might not be awarded according to political criteria). Progress on the political front (toward limited democracy) would also come to a halt -- indeed, would be reversed.^[37]

Post-1997 developments in Hong Kong are inevitably subject to conflicting interpretations, and achieving a meaningful synthesis is not a realistic objective at this juncture. To complicate the picture, China is not the sole factor in the equation, the implication being that other factors (e.g., an insecure and over-anxious local government which is prone to “second guess” its ultimate “controllers” in Beijing) are influencing policy outputs. Nevertheless, it is legitimate to argue that the status quo seven years following the transfer of sovereignty is clearly more palatable than suggested by the pre-1997 forecasters and there is no compelling reason to conclude that the outlook is likely to deteriorate significantly in the foreseeable future.

On the negative side, China has impeded, at times in a distinctly heavy-handed style, progress toward representative democracy.^[38] By the same token, it has apparently provided the inspiration for a highly restrictive national security legislation (shelved for the time being, after provoking a massive backlash across the social spectrum).^[39] On the positive side, Hong Kong has continued to enjoy a fairly high degree of autonomy in both the domestic and external arenas. The local government cannot be said to have used its authority as effectively as the circumstances obviously warrant and its performance shows few signs of “learning-by-doing.”^[40] Be that as it may, the Beijing “proxies” in Hong Kong have taken no concrete steps to curtail freedoms in the academic, communications, economic, legal, or other strategic domains.

As one surveys the current scene at a more fundamental level, the sense of unease stemming from the encounter between the model resorted to for purposes of prognostication, and the behavioural patterns actually observed, increases further. Model-building entails a measure of (over)simplification so that the emphasis here is not on the discrepancy per se but its magnitude. Game theorists see a cluster of self-interested groups in China whose unambiguous goal is to marginalise Hong Kong, which have all the means necessary to pursue this problematic objective, and whose leaders will consistently and systematically operate in accordance with that logic.

The present-day flow of actions -- in Beijing and elsewhere in the mainland, let alone in Hong Kong -- scarcely corresponds to this portrayal. Quite often, alliances are loose and variable, goals are ill-defined and in a state of flux, means are diffuse and poorly mobilised, and issue management is reactive and arbitrary.^[41] To illustrate, China occasionally seeks, in a seemingly irrational way, to tighten the grip on Hong Kong, without carefully considering the ramifications for Taiwan (its conduct in Tibet is, on the face of it, perhaps even less productive in this respect).

The gap between the pre-1997 predictions and today’s realities is thus sufficiently large to justify at least a partial examination of the highly formal approach employed before the transfer of sovereignty to depict, in broad terms, international legal behaviour well beyond that potentially disruptive event.

IV. A Cautionary Tale

Game theory is a powerful analytical tool which can bring into sharp focus key dimensions of complex problems and play a useful role in exploring them in a dynamic fashion. It can also serve as an effective vehicle for seeking solutions to such problems. Both as a descriptive and normative research instrument it has few rivals in terms of versatility. Enforcement theorists and other legal scholars have amply demonstrated that it can be productively applied in the traditionally qualitative and philosophically-oriented field of international law/international legal compliance. Researchers employing a similar conceptual/methodological apparatus have shown that, in principle, it can be brought to bear on thorny questions such as China’s future actions regarding Hong Kong, a territory with which it is externally-bound to conduct its affairs in a predetermined manner.

The purpose of this paper is not to evaluate, critically or otherwise, the results of this particular case study. As indicated earlier, it has been chosen in order to illustrate in a specific geographical/historical context, with which the authors are well-familiar, that game theorists practice what they preach, at times boldly so. The preceding sections are intended to provide a backdrop against which an examination of the limitations of game theory as a tool of international legal analysis could be meaningfully undertaken. Again, since the focus is on the cons rather than the pros, a reminder is needed that this is a “rebalancing act” rather than a “balancing” one.

Perhaps the most disconcerting feature of analytical efforts, such as that addressing the uncertainty stemming from the resumption of Chinese sovereignty over Hong Kong is, the ease with which those engaged in them construct the framework for drawing systematic conclusions about the results of macro-level social interaction in such complex circumstances. The tasks performed are very challenging [again, to specify/produce: (1) the identity of the relevant internal and external actors who may wish to influence the policy in question; (2) a set of policy alternatives that encompasses all conceivable patterns; (3) the policy preferences of each group on the issue; (4) estimates of the economic, physical (not just military), political, and social capabilities that each group may employ to influence the policy decision; and (5) estimates of the importance (salience) each group attaches to each issue, signifying the group’s interest in influencing policy outcomes]. Yet, the impression given is that this is a relatively simple undertaking, involving quasi-mechanical progression.

In fact, difficult choices confront the analysts at each stage of the exercise. There is no guarantee that two persons, no matter how close theoretically and methodologically, would proceed in a similar fashion from the point of departure to the final destination. Indeed, one cannot rule out the possibility of the same analyst following a markedly different path when revisiting the same problem at a subsequent juncture. The claims of scientific rigour notwithstanding, game-like explorations in international legal settings are thus open to the criticism that they lack “reliability.”^[42] The corollary is that it behoves those who embrace the theory/method to formulate “guidelines”, at least tentative in nature,^[43] for placing such endeavours on a more solid footing. Otherwise, they could legitimately be viewed as parochial and of limited scientific/practical (as distinct from heuristic) value.

The tendency of game theorists to operate in a solo manner when dissecting international issues with legal dimensions is a manifestation of the procedural inadequacies characterising their approach. More often than not, they are content to examine multifaceted situations requiring inputs from several disciplines and diverse methodological sources on their own. Post-1997 Hong Kong/China realities certainly need to be assessed by researchers capable of observing the regional dynamics through a wide range of analytical devices and whose skills extend beyond mathematical economics. This may include traditional-style legal scholars, historians, geographers, political scientists, sociologists, and economists (e.g., institutional and policy-oriented) less inclined to rely on quantitative tools for generating strategic insights.

Another possible criticism relates to choices made regarding the goals (or, to express it more narrowly, policy preferences) of actors in the international arena. Game theorists assert that such actors are “utility-maximisers,” but that the term is free of any particular connotations. A group may thus be driven by “altruistic” (as distinct from “selfish”) factors without necessarily violating the logic underlying this form of analysis. The idea is by no means unreasonable and one does not have to subscribe to the management approach to international legal compliance to embrace it. The notion can comfortably be accommodated within a framework firmly based on the principle of utility maximisation.

In practice, however, self-interest almost invariably overshadows all other goals. The Chinese groups identified as relevant vis-à-vis Hong Kong are no exception to the norm. The centralisers, decentralisers, and military officers pulling the strings in Beijing and the power centres in the provinces apparently have no concern whatsoever for their compatriots in the former British colony. Rather, they are selfishly bent on extracting as much advantage for themselves, stripping Hong Kong residents of most fundamental economic and political rights in the process. Needless to say, this picture, even if meticulously painted, is difficult to reconcile with reality (let alone alternative theories), and perhaps defies common sense.^[44]

Goals themselves are neither “rational” or “irrational,” and game theorists possess no special qualifications to designate them as such. Rationality, or lack thereof, can be established, although this is a serious undertaking, by studying the relationship between goals and environmental opportunities/constraints or, better still, between goals and means. Game theorists assume that actors in the international arena choose goals that are “best,” given available environmental opportunities/prevailing environmental constraints, or that they choose the most appropriate means to satisfy their goals. This is a convenient assumption which greatly simplifies the analytical task. Its validity may nevertheless be debated, whether selectively or across the entire decision spectrum.^[45]

The goals-opportunities/constraints/means nexus remains generally obscure in applications of game theory to international legal problems. It is not entirely clear how the Chinese groups supposed to be ill-disposed toward Hong Kong, or likely to benefit from its fall from grace, are expected to execute their designs regarding the capitalist enclave. Unless the strategies are carefully outlined, there is insufficient transparency to evaluate the rationality of actions in this specific case.^[46] The pre-1997 analysis of post-1997 international legal compliance in the Hong Kong/ China context cannot unfortunately be dismissed as an aberration. Such “errors of omission” are the inevitable upshot of “reductionist” approaches to group conflict and cooperation.^[47]

The preoccupation with the groups may give rise to another criticism. Specifically, stretching the concept too far is bound to lead to a configuration characterised by blurring of key analytical boundaries (amounting at times to a veritable “embarrassment of riches”). Groups tend to proliferate in such circumstances and it is not easy to distinguish between those that qualify as “central” and those that are “peripheral” in nature. Perhaps more importantly, there is an inherent risk of losing sight of the fact that the state -- an entity subject to intense group pressure but with a distinct identity of its own -- is the principal actor in the international arena. “Realists”, whether of the “classical” or “structural” variety, may accord excessive attention to the state, yet game theorists often swing wildly in the opposite direction.^[4]

It should also be noted that the view that decision makers, at all levels (international, domestic, macro, micro, etc.), invariably seek to make “optimal” choices when matching goals with opportunities/constraints/ means is not universally shared. This is another case of an “embarrassment of riches” because the theoretical universe is very large. Rather than explore it thoroughly, one may simply point out that there has been a decisive shift in the past twenty years or so away from the rational choice model in its classical form (positing that decision makers have knowledge of all alternatives, their consequences, and a consistent preference ordering coupled with a decision rule that allows an unambiguous identification of the “best” course of action) toward the notion of “bounded rationality.”^[48]

The latter is grounded in empirical research, both of the experimental and field varieties, highlighting the cognitive and institutional constraints facing decision makers in complex social settings. Given the prevalence of such constraints, actors in the public arena more often than not pursue a “satisficing” path, or settle for performance that is merely “good enough” rather the best possible in the circumstances (this pattern is not confined to the public arena, but also manifests itself in human behaviour in less elaborate contexts because of limitations stemming from the inability of the cognitive apparatus to process information, particularly on a large scale and of multifarious types, in an efficient manner).^[49]

From a bounded rationality perspective, China’s compliance with its legal obligations vis-à-vis Hong Kong is a subject requiring a broader research agenda than that constructed by game theorists. The assumption that policy action in an environment where strong domestic and international interests converge is the exclusive product of manoeuvres by three fairly homogenous groups, coherently and consistently driven by a narrowly-based cost-benefit calculus, cannot be embraced with a degree of confidence. Interestingly, an attempt to address the issue from a management approach standpoint would have probably also resulted in a one-dimensional configuration, bearing limited correspondence to a highly complex political reality.^[50]

The extent to which rationality manifests itself in the public arena inevitably varies with the conditions affecting the ability of the relevant actors to operate effectively and exercise full discretion. For example, some decision theorists contend that rational “search” is possible when the prevailing constraints are at the low end of the feasible range and the subsystem within which specific policies originate is decidedly simple (Table 1). This is merely one of several competing typologies, but they all lead to generally similar conclusions. Clearly, the formation and maintenance of international law involves serious constraints and great complexities (even in the Hong Kong/China case, it would be difficult to challenge the argument that they qualify as no less than “moderate”). Strict adherence to the rationality principle may thus not be the optimal research strategy.^[5]

The distinction between the formation of international law and its maintenance is useful in this context. The reason lies in the fact that the former is more likely than the latter to bring into play rationality in its classical form. After all, on balance, negotiating international treaties involves higher stakes than complying with them. The corollary presumably is that it calls for a greater concentration of strategic effort, minimum fragmentation of the decision-making structure, unambiguous definition of the policy goals, and careful scrutiny of the available means to be deployed in pursuit of the ends agreed upon. Again, the inference to be drawn is that rationality is a variable that depends on a host of situational influences.^[52]

The distinction between the formation and maintenance of international law corresponds in many respects to that between strategy formulation and strategy implementation. This has further implications for the efforts of game theorists to direct their analytical tools toward the rule-based dimension of international relations. The burgeoning literature in the policy studies field suggests that, other things being equal, strategy formulation is a more deliberate process than strategy implementation. It follows that international legal compliance, which largely consists of the execution of established policies, cannot be approached exclusively from a perspective rooted in the notion of rationality, as traditionally conceived.^[5]

The apparent relevance of the special circumstances in which policies are evolving, as well as the distinction between the various phases of the entire process (the one between strategy formulation and strategy implementation being the most obvious), lends credence to the “depth-of-cooperation” hypothesis in that it reinforces the proposition that all treaties are not equal, in terms of the burden they impose on the signatories and the challenge they present with respect to compliance. The argument may evidently also be embraced in reverse, or from the back end, because if international agreements do indeed vary in their depth, they may elicit different responses from policy makers. Those which carry considerable strategic weight may command their full attention, and those which perhaps do not (e.g., the Sino British Joint Declaration, as viewed from Beijing) may be relegated to the policy periphery. The latter may not provide game theorists, firmly wedded to the pure idea of rationality, with fertile ground for their explorations.^[5]

By definition, games feature interaction between players. The theorists who identify and profile them write the script as well and -- given the nature of the enterprise, particularly the strict assumptions underlying the models employed -- execute the task in a distinctly static manner. The players and their profiles do not change meaningfully over time. Nor does the script (e.g., the “rules of the game” and the environment in which they are embedded may not vary materially during periods characterised by profound structural shifts). The same domestic groups in China thus continue -- in the face of internal and external “shocks,” both of the negative and positive variety -- to pursue the same strategies vis-à-vis Hong Kong. The curtain falls on “old eras” and “new eras” are born; one century draws to a close and another comes into being. Yet, this complex picture remains intact -- needless to say, unrealistically so.^[54]

The typical game has a deceptively simple structure. In one example, reasonably representative of the genre, there are two players who confront two choices familiar to students of international legal compliance: co-operate or defect (Table 2). Depending on the other party’s response, each course of action generates utility/disutility for the players (quantifiable, predetermined, and constant). Such a game normally has an equilibrium, or an outcome that the players should seek in light of the behavioural postulates of the rational choice/expected utility model. In the specific example used here for purposes of illustration, and other similar settings, whether real or hypothetical, the concept of equilibrium serves a crucial analytical function by providing game theorists with the basis for prediction. In the other words, by pinpointing the supposed equilibrium of the Hong Kong-China legal compliance tangle, or its equivalent, the pre-1997 researchers have been able to extrapolate trends well beyond the transfer of sovereignty.^[55]

The concept of equilibrium is at present inextricably linked to the name of John Nash, the Nobel laureate whose painful mental journey, extending over his entire professional life, has been portrayed vividly in the movie A Beautiful Mind. The brilliant economist/mathematician has devised a method of solving games that is appropriately called a “Nash equilibrium,” a no-regrets outcome in which all players are satisfied with their strategy, given the actions of all the other parties involved. In such a situation, one is not necessarily pleased with the strategies pursued by the latter. Rather, the course of action chosen is an optimal response to the opponents’ moves.^[56]

A simple example (labour law-oriented, rather than international law-related!) may illustrate the underlying logic. The “game” features two employees, Abe and Benny, who both seek a salary increase. It is convenient to assume that if just one of them takes the initiative, he would be successful, but if both do, their employer would react adversely and proceed to dismiss them. This game apparently has two Nash equilibria: one where only Abe asks for a raise and one where Benny asks for one. An outcome cannot be considered as a Nash equilibrium if neither employee seeks a salary increase, because each would “regret” not taking the step, knowing that the other did not do so. By the same token, an outcome cannot qualify as a Nash equilibrium if both request a raise, since each would then “regret” being fired.

The attraction of a Nash equilibrium stems from its stability, and that is true of the entire genre rather than merely this specific variant. Nevertheless, problems proliferate when sufficiently realistic international legal scenarios are squeezed into the analytical straitjacket. As is commonly the case, information is available in abundance, resource (e.g., time) scarcity is not an issue, external and internal pressures are conspicuous in their absence, the players are engaged in a game with serious implications yet they do not cooperate, and no matter how formidable their skills are, altering each other’s strategies is not a feasible option. These reservations surface in many contexts and attention hence needs to be drawn to the fact that games, certainly those which are played in the international arena, normally have several equilibria. To the extent that none is inherently “superior”, the whole system is obviously “indeterminate”, which renders meaningful prediction (or for that matter, description) a distinctly elusive task.^[57]

The idea that the complex and dynamic Hong Kong-China legal compliance “game” has a single equilibrium -- or even several, but one which can be characterised as “superior” in terms of credible policy criteria -- is difficult to embrace on practical grounds. In addressing such an analytical challenge, it would arguably be preferable to proceed in an open-ended manner and rely on qualitative research instruments, albeit within a coherent methodological framework and along a path conducive to achieving a reasonable degree of accountability. Scenario writing may serve as an effective tool of inquiry when one operates across a wide thematic spectrum and over a long period of time. It is an academic art, rather than a science in the strict sense of the term, geared toward generating hypothetical sequences of events constructed for the purpose of focusing attention on causal processes and decision points (Figure 1 - below).

As the best-known practitioner of the craft has elaborated: “scenarios… answer two kinds of questions… (1) Precisely how might some hypothetical situation come about, step by step? and (2) What alternatives exist, for each actor, at each step, for preventing, or facilitating the process?”^[58] This is by no means the only research vehicle available for exploring the complexity and dynamic shifts associated with international legal compliance.^[59] Nevertheless, it is appropriate to highlight both its features and merits here, because it contrasts sharply with the restrictive and abstract - yet increasingly fashionable - approach examined in this paper.

V. Conclusion

Game theorists have broadened their horizons significantly in the past twenty years or so and are currently active in virtually every branch of the social sciences. The law, which may qualify as such for analytical purposes, has been no exception. On the domestic front, the involvement has assumed major proportions^[60] and there has been notable progress, albeit more tentative and uneven, on the international side.^[61] Many insights -- theoretical, methodological, and of the strategic variety -- have been produced and it is legitimate to contend that the heuristic value of the work undertaken has been considerable. Game-type logic can be instrumental in uncovering obscure dimensions of an elaborate picture, bringing academic/

policy/everyday problems into a clear focus, and exposing weaknesses of diagnostic/prognostic/prescriptive research resting on “softer” foundations.^[62]

It would be tempting, in light of the preceding discussion, to conclude that unfortunately the benefits do not extend any further, insofar as international law is concerned. While the Hong Kong-China relationship undoubtedly reveals the limitations of game theory as an instrument for dissecting international legal compliance, there is no compelling reason, on this basis, to adopt such an unequivocal stance. Rather, it may be appropriate to suggest that this form of analysis be confined to the “appropriate domain” (e.g., the formation of international law versus its maintenance; “segmented universalism”) and be integrated with other theoretical/methodological/practical approaches in an ad hoc/eclectic/

opportunistic manner (“partial universalism”). Otherwise, it will sooner or later recede into the background.^[63]

Game theorists are not entirely oblivious to the criticisms levelled at them. Indeed, they have sought to address the reservations expressed partially, without diluting the essence of their creed. The notion of “nested games” thus ingeniously allows for the possibility of an actor consciously not opting for the “best” response when confronting a series of choices, yet operating within an overall “rational” framework.^[64] By the same token, the “theory of moves” elegantly injects a modicum of flexibility and dynamism into an overly rigid and static structure.^[65] These are interesting developments that should be welcomed by students of international legal compliance and other scholars.^[66] Nevertheless, they do not go far enough to obviate the need for selectivity and fusion with other approaches.

TABLES AND DIAGRAMS

Source: Adapted from S. Brooks and A. G. Gagnon, The Political Influence of Ideas (New York: Praeger, 1994).

Source: Adapted from J. E. Lane and S. Ersson, Government and the Economy (London: Continuum, 2002).

Figure 1

Kahn’s View of Scenario Construction

Source: Adapted from H. Kahn and A. J. Wiener, The Year 2000 (New York: Macmillan, 1967).

^[*] Miron Mushkat, Visiting Professor, Department of Politics and Public Administration, University of Hong Kong. Roda Mushkat, Professor and Head, Department of Law, University of Hong Kong.

^[1] Khun TS, The Structure of Scientific Revolutions (Chicago: University of Chicago Press, 1962); and Horwich P (ed), World Changes (Cambridge: MIT Press, 1994).

^[2] Chayes A and Handler Chayes A, The New Sovereignty (Cambridge: Harvard University Press, 1995).

^[3] Young OR, ‘The Effectiveness of International Legal Institutions’ in Rosenau JN and Czempiel EO (eds), Governance without Government (Cambridge: Cambridge University Press, 1992) 183.

^[4] Chayes and Handler Chayes, above n 2.

^[5] Ibid, 11.

^[6] Ibid.

^[7] Ibid, 22.

^[8] Ibid, 32-33.

^[9] Ibid.

^[10] Ibid.

¹⁰ Ibid.

^[11] Ibid.

^[12] Olson M, The Logic of Collective Action (Cambridge: Harvard University Press, 1965); Axelrod R, The Evolution of Co-operation (London: Penguin, 1984); Oye KA (ed), Cooperation Under Anarchy (Princeton: Princeton University Press, 1986): Yarbrough BV and Yarbrough RM, Cooperation and Governance in International Trade (Princeton: Princeton University Press, 1992); and Bayard TO and Elliott K, Reciprocity and Retaliation in U.S. Trade Policy (Washington: Institute for International Economics, 1994).

^[13] Downs GW, Rocke DM, and Barsoom PN, ‘Is Good News about Compliance the Good News about Cooperation?’ (1996) 50(2) International Organization 379-406; Dorn WA and Fulton A, ‘Securing Compliance with Disarmament Treaties’ (1997) 3(1) Global Governance 17-40; and Downs G, ‘Enforcement and the Evolution of Cooperation’ (1998) 19(2) Michigan Journal of International Law 319-344.

^[14] Haas PM, ‘Compliance with EU Directives’ (1998) 5(1) Journal of European Public Policy 19.

^[15] Young OR, Compliance and Public Authority (Baltimore: Johns Hopkins University Press, 1979); and Underdal A, ‘Explaining Compliance and Detection’ (1998) 4(1) European Journal of International Relations 5-30.

^[16] Stein AA, ‘Coordination and Collaboration’ in Krasner SD (ed), International Regimes (Ithaca: Cornell University Press, 1983) 1-27.

^[17] Downs, Rocke and Barsoom, above n 14, 385; Axelrod and Keohane, above n 13; Dorn and Fulton, above n 14; and Underdal, above n 16.

^[18] Downs, Rocke and Barsoom, above n 14.

^[19] Ibid; and Tallberry J, ‘Paths to compliance’ (2002) 56(3) International Organization 609-643.

^[2]1 Mushkat R, One Country, Two Personalities -- the Case of Hong Kong (Hong Kong: Hong Kong University Press, 1997).

²⁰ Mushkat M, The Economic Future of Hong Kong (Boulder: Lynne Reinner, 1990).

^[21] Ghai Y, Hong Kong’s New Constitutional Order -- the Resumption of Chinese Sovereignty and the Basic Law (Hong Kong: Hong Kong University Press, 1999).

^[22] Mushkat R, above n 21.

^[23] Dai X, ‘Information Systems in Treaty Regimes’ (2002) 54(4) World Politics 405-436.

^[24] De Mesquita BB, Newman D and Rabushka A, Forecasting Political Events (New Haven: Yale University Press, 1985); and De Mesquita BB, Newman D and Rabushka A, Red Flag over Hong Kong (Chatham: Chatham House, 1996).

^[25] This was attempted in another publication, upon which this paper partly draws: Mushkat M and Mushkat R, ‘The Political Economy of International Compliance’, (2004) 10(2) University of California Davis Journal of International Law and Policy 229.

^[26] De Mesquita, Newman, and Rabushka, Red Flag over Hong Kong, above n 26.

^[27] Ibid.

^[28] Ibid.

^[29] Ibid.

^[30] Ibid.

^[31] Ibid.

^[32] Ibid.

^[33] Ibid.

^[34] Ibid.

^[35] Ibid.

^[36] Ibid.

^[37] Ibid.

^[38] Mushkat M and Mushkat R, ‘The Political Economy of Constitutional Reform in an Externally-Constrained Environment’, Journal of Malaysian and Comparative Law (forthcoming).

^[39] Mushkat M and Mushkat R, ‘The Political Economy of Constitutional Conflict in Hong Kong’ (2004) 11(4) Tillburg Foreign Law Review 756-781.

^[40] Lo S, Governing Hong Kong (New York: Nova Science, 2001).

^[41] Mushkat M, ‘The Political Economy of China’s Future’ (2002) 31(3) Global Economic Review 41-66; and Mushkat M, ‘Economic Reform, Discontinuous Change, and China’s Future’ (2003) 2(2) Perspectives on Global Development and Technology 255-282.

^[42] Neuman WL, Social Research Methods (Boston: Allyn and Bacon, 2000); and Schutt RK, Investigating the World (Thousand Oaks: Pine Forge Press, 2001).

^[43] Bryman A, Social Research Methods (Oxford: Oxford University Press, 2001); and Bernard HR, Research Methods in Anthropology (Walnut Creek: Altamira Press, 2002).

^[44] Mushkat and Mushkat, ‘The Political Economy of International Legal Compliance’, above n 27.

^[45] Brams SJ, ‘Game Theory’ (2000) 1(3) International Studies Perspectives 221-232; Jones BD, Politics and the Architecture of Choice (Chicago: University of Chicago Press, 2001); Munck GL, ‘Game Theory and Comparative Politics’ (2001) 53(1) World Politics 173-204; Brams SJ, Biblical Games (Cambridge: MIT Press, 2002); and Brams SJ, Negotiation Games (New York: Routledge, 2003).

^[46] Mushkat and Mushkat, ‘The Political Economy of International Legal Compliance’, above n 27.

^[47] Bryman, above n 45; and Bernard, above n 45.

^[48] Jones, above n 47.

^[49] Ibid.

^[50] Mushkat and Mushkat, ‘The Political Economy of International Legal Complaince’, above n 27.

^[5]3 Keohane RD, After Hegemony (Princeton: Princeton University Press, 1984).

⁵¹ Ibid.

^[52] Brooks and Gagnon, above Table 1; and Thompson JD, Organizations in Action (New York: McGraw Hill, 1967).

^[5]6 Hrebiniak LG and Joyce WF, Implementing Strategy (New York: Macmillan, 1984); and Hill M and Hupe P, Implementing Public Policy (London: Sage, 2002).

⁵³ Mushkat and Mushkat, ‘The Political Economy of International Legal Compliance’, above n 27.

^[54] Ibid.

^[55] Ibid.

^[56] Dixit A, Games of Strategy (New York: Norton: 1999); Watson J, Strategy (New York: Norton, 2002); and Miller JD, Game Theory at Work (New York: McGraw-Hill, 2003).

^[57] Brams, ‘Game Theory’, above n 47; and Munck, above n 47.

^[58] Kahn and Wiener, above Figure 1.

^[59] McCune SD, Guide to Strategic Planning for Educators (Alexandria: Association for Supervision and Curriculum Development, 1986); Twiss, B. C., Forecasting for Technologists and Engineers (London: Peregrinus, 1992).

^[60] Baird DG, Gertner RH, and Posker RC, Game Theory and the Law (Cambridge: Harvard University Press, 1999); and Meyerson MI, Political Numeracy (New York: Norton, 2002).

^[61] Krasner, above n 17.

^[62] Jones, above n 47; Hollis M and Nell E, Rational Economic Man (Cambridge: Cambridge University Press, 1975); Monroe KR, The Economic Approach to Politics (New York: Cambridge University Press, 1991); Green DP and Shapiro I, Pathologies of Rational Choice (New Haven: Yale University Press, 1994); Hollis M, Trust Within Reason (Cambridge: Cambridge University Press, 1998); and Hay C, ‘Theory, Stylized Facts, or Heuristic Self-Fulfilling Prophecy? The Status of Rational Choice Theory in Public Administration’ (2004) 82(1) Public Administration 39-62.

^[63] Brams, ‘Game Theory’, above n 47; and Munck, above n 47.

^[64] Tsebelis G, Nested Games (Berkeley: University of California Press, 1990).

^[65] Brams SJ, Theory of Moves (New York: Cambridge University Press, (1999).

^[66] See also: De Mesquita BB, Predicting Politics (Clumbus: Ohio University Press, 2002); and De Mesquita et al, Political Survival (Cambridge: MIT Press, 2003).